[crossfire] Rebalancing, difficulty curve -- simple ideas

[Mark Wedel]

Alex Schultz wrote:
> Lalo Martins wrote:
>> Ok, one issue with balance is that, as someone mentioned, most monsters
>> either kill you too easily or are killed too easily.  I don't have a
>> solution for that, only wanted to bring it up so we don't forget it ;-)
>>   
> Well, I'm not really sure, but I think there are two main causes for that:
> -The of chance of hitting isn't very random, it's AC and WC, and has
> some random variation, however the random variation seems extremely
> insignificant compared to AC and WC.
> -The speed of hits is very fast, so the random variation of AC vs. WC
> gets averaged out very quickly.
> We could try to reduce the speed of battle, however that would take ALOT
> of tweaking to balance and not frustrate players. Perhaps one option
> would be to make it so the random variation of AC vs. WC should last
> longer than one hit, and "ebb and flow" a bit.

  Thinking about this, there are several issues:

  Even at level 1, a fighter type character will have a weapon speed of ~1.5 and 
a damage of 10.  This means that each time he hits, he will average 8 points of 
damage

  kobolds have 2 hp.  Pretty much no matter what you have, you will kill them.

  goblins have 6 hp.  More or less same thing.

  orcs have 4 hp.

  gnolls have 8 - likely to take two hits.

  So the issue right there is that these creatures do not have many hp.

  In terms of their speed, the fastest of that group are orcs with a speed of 
0.15 - and orcs do dam 1.

  gnolls have speed 0.1, dam 4.

  So at least at low level, that pretty clearly shows that things are geared for 
monsters being much easier to kill than players.

  Fixing this is harder.  Various thoughts (note that all of these would have to 
be targeted towards a major release):

1) Reduce weapon speed - in fact, I'd suggest that weapon speed be driven purely 
by the weapon.  This fixes some other problems, namely huge amounts of damage 
players inflict at high levels - not only do players get very high weapon speed, 
  but they start doing a lot more damage, so effectively, at high level, a 
character is probably doing 10 times more damage than at low levels.  Sure - 
they should do more, but not sure if 10x is the right answer (and in this case, 
high level may be ~30)

2) Increase players starting hp - in this way, for these monsters, players could 
take more hits and still survive.

3) At least at low levels, these monsters may need to be toughened up.  In order 
to not muck too much with spells, perhaps give them a better natural armor so 
damage players cause is reduced (but I'd probably suggest that hp be increased)

4) Maybe change ac/wc?  Right now, it is a d20 system, so any changes are big - 
a gnoll has AC13 and wc 12 - and in fact, all the monsters above have a 10+ ac, 
and my first level test character has a wc of 18.  The basic hit method is if 
(ac + d20) > wc.  So for the best of the creatures, with and ac of 13, it means 
that my character hits on a 5+, or 75% of the time.

  If we change it to a percentile system, that may have an advantage of greater 
variation, especially if the increase rate is reduced (1%/skill level?)  Why 
that makes a difference is this - if you are in a marginally difficult to hit 
situation (say need a 15+), it means that having a few objects that give you a 
wc bonus, or being a few levels higher and having more bonuses could double how 
often you hit (10+ on a d20)

  This makes it very difficult to tune - one player could run through, and say 
'that was too easy - it needs a higher ac'.  Next person finds it impossible, 
because they just can't hit it.

  IF instead, it is a 50 vs 70, that is a ~30% difference - a big gap, but 
certainly a big difference from infrequently hitting to hitting all the time.


  With all that said, other thoughts:
Each level range should undergo testing.  Eg, concentrate on testing levels 1-5 
(lets say) until all the balance there is worked on, then starting testing 5-10, 
10-20, etc.

  This is because these changes will determine character to some extent - how 
hard it is to get to level 10 is likely to determine the items the character has 
at level 10, how many times they have improved stats, etc.

  I'm not sure that ebbing/flow attacks would really make any difference - it 
just means that when your hot, you kill a bunch of things quickly, when it goes 
the other way, you don't kill anything - on average, it doesn't seem like this 
would make much a difference, it would just seem frustrating to players (I was 
able to kill these really quickly an hour ago!).



>> To encourage player interaction, the game needs to strongly nudge
>> characters into having different strengths and weaknesses.  This is
>> already done to some extents (at low levels) by resistances, natural and
>> god-given; of course at higher levels you start collecting artifacts and
>> you have any resistances you want.
>>   
> Personally I think we should also consider ways to make different
> classes slightly better at different skills. Perhaps a "skill attument"
> of sorts. Also class stat bonuses are currently very pointless very quick.

  I think some of this has been discussed before.  Different type of skills, 
which determine rate of exp.

  The entire stat thing should probably be rethought - with current stat cap, 
stat potions become useless.  More variation based on the starting stats could 
make sense.

  If we go to a 1-100 stat values, and say 20 is the nominal starting stat, with 
say ±5 based on race, you could make max stat based on 5*modifier.

  Lets say normal max stat is 50.  If your character has a +5 strength bonus, 
his max strength is 75.  If you have a -5 int, max int is 25.

  OR something like that - that would make a lot bigger difference on those 
modifiers.  Right now, it is typically just ±2 or so, which has that same affect 
on max stat - it basically means that all the character stats are pretty close, 
and with magic, they can pretty easily be all the same.

  (under such a changed stat system, the bonuses most items give would also have 
to be increased to some extent)

> 
>> An idea I think might be worth experimenting with is cutting the
>> contribution that *ALL* skills make to general score.  If you have more
>> characters running around with level ~10 and their primary skill(s) at ~20
>> that would encourage them to get good at one skill, and therefore to
>> interact with other players.
>>   
> IMHO this would be a good idea, though one would have to be careful
> about how much one cuts it by. Perhaps how much it cuts it by should
> also vary on a curve depending on the level.

  I'm unsure how this makes any difference, unless you can somehow identify some 
skill as primary skill.

  Otherwise, all it seems you are doing is reducing exp gain - if I get 500 exp 
for literacy vs 500 exp for 1 handed weapons, how does that affect what gets 
contributed to overall score/exp?

  If you really want to change this, one could just remove skillscrolls from the 
game - what you start with is all you get.

  There are certainly issues that each character is basically standalone and 
don't need the others - I'm not sure if that can really be forced upon players. 
  AT some level, that has to be allowed - if no one is on, I need some way to be 
able to heal myself, identify items, etc.

  Slowly combat down a bit may help out a bit - right now, most things are so 
fast that it can be really difficult for players to coordinate - if I kill that 
creature in 5 seconds, it becomes difficult to have the archer player get behind 
it an pelt it with arrows, etc (by the time the archer gets there, it is 
probably dead)

> 
>> Finally, the easy answer to the curve issue is tweaking the exp_table
>> carefully.  I want to start a mini-project, with the goal of tuning up a
>> "perfect" (heh) exp_table.
> Well, I don't think tweaking the exp_table is the 'silver bullet' that
> will fix it the whole curve issue, however I feel it is an important
> step towards fixing it.

  I would say that it probably needs to figured out what are the problems that 
need to get fixed first, or what all will be fixed/changed.

  If combat is slowed down, you probably want to wait on adjusting the exp table 
until that is done - simply because you might find that with slower combats, 
characters get exp slower, etc.


  One other thought on this - the spells should be rebalanced/redistributed.

  IIRC, crossfire originally only had 30 levels.  This is why for the most part, 
the top spell level is about 20.

  With 100 levels, this doesn't make a lot of sense.  Rebalancing spells is a 
bit of work (as if you increase the level you get the spell, you probably need 
to increase damage a bit, etc).

  But back in terms with cooperative play in mind, we may want to also focus 
more on spells which work good in groups.  In particular, things like cones and 
exploding ball spells probably are not good - bullets, bolts, summoned 
creatures, healing, protective, make sense